Impact of chronic kidney disease and end-stage renal disease on outcomes after complex endovascular and open aortic aneurysm repair

Objective: Chronic kidney disease (CKD) and end-stage renal disease are traditionally associated with worse outcomes after endovascular aortic repair (EVAR) and open aneurysm repair (OAR) of abdominal aortic aneurysms (AAAs). However, there needs to be more data on complex AAA repair involving the aorta's visceral segment. This study stratifies complex AAA repair outcomes by CKD severity and dialysis dependence. Methods: All patients undergoing elective OAR and fenestrated/branched EVAR (F-BEVAR) for complex AAA with preoperative renal function data captured by the Vascular Quality Initiative between January 2003 and September 2020 were analyzed. Patients were stratified by CKD class as follows: normal/mild (CKD 1 and 2), moderate (CKD class 3a), moderate to severe (CKD 3b), severe (CKD class 4 and 5), and dialysis. Only patients with clamp sites above one of the renal arteries were included for complex OAR. For F-BEVAR, patients with proximal landing zones below zone 5 (above celiac artery) were included, and distal landing zones between zones 1 and 5 were excluded. Primary outcomes were perioperative and 1-year mortality. Predictors of mortality were identified by Cox multivariate regression models. Results: We identified 7849 elective complex AAA repairs: 4230 (54%) complex OARs and 3619 (46%) F-BEVARs. Most patients were White (89%) and male (74%), with an average age of 72 ± 8 years. The patients who underwent F-BEVAR were older and had more comorbidities. Elective F-BEVAR for complex AAA started in 2012 and increased from 1.4% in 2012 to 58% in 2020 (P < .001). The OAR cohort had more perioperative complications, but less 1-year mortality. The normal/mild CKD cohort had the highest 1-year survival compared with other groups after both complex OAR and F-BEVAR. On Cox regression analysis, when compared with CKD 1-2, worsening CKD stage (CKD 3b: hazard ratio [HR], 2.5; 95% confidence interval [CI], 1.82-3.40; P < .001; CKD 4-5: HR, 1.9; 95% CI, 1.16-3.26; P = .011; and dialysis: HR, 4.4; 95% CI, 2.53-7.72; P < .001) were independently associated with 1-year survival after F-BEVAR. After complex OAR, worsening CKD stage but not dialysis was associated with 1-year mortality compared with CKD 1-2 (CKD 3b: HR, 1.6; 95% CI, 1.13-2.35; P = .009; CKD 4-5: HR, 3.4; 95% CI, 2.03-5.79; P < .001). Conclusions: CKD severity is an essential predictor of perioperative and 1-year mortality after complex AAA repair, irrespective of the treatment modality, which may reflect the natural history of CKD. Consideration should be given to raising the threshold for elective AAA repair in patients with moderate to severe CKD and end-stage renal disease, given the high 1-year mortality rate.